Section 3: Improvements to police-reported crime data
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Counting police-reported crimes
In the late 1980s, Statistics Canada began collecting in-depth information about each criminal incident coming to the attention of police through a new version of the Uniform Crime Reporting Survey known as the UCR2. Police services across the country have gradually been adopting this new survey as their records management systems have become capable of collecting and providing this level of detail.
By 2007, virtually all police services in Canada were providing incident-based crime data on the nature of criminal incidents, accused persons and victims. There are now nearly 200 separate criminal offence categories covered by the UCR2.
The greater availability of incident-based data has made it possible to address some of the limitations associated with the way particular criminal offences have historically been counted. Two offences, namely counterfeiting and robbery, have been most affected. A summary of these changes and their impacts are outlined below. Further details on these changes and those made to other offence counts can be found in Appendix B of this report.
Counterfeiting
The purpose of the UCR2 survey is to collect data on crimes reported to and substantiated by police services in Canada. For much of the past decade, Statistics Canada has been collecting counterfeiting data not only from police services, but also from the RCMP's Bureau of Counterfeit and Document Examinations.
Recent experience has shown that many of the incidents coming from the Bureau of Counterfeiting and Document Examination have likely been detected by merchants or banking institutions following a financial transaction. These incidents have neither come to the attention of, nor been substantiated by, police services.
Further, in many of these incidents, counterfeit bills have been passed unwittingly by individuals. It is important to note that passing a counterfeit bill without knowing it is not genuine does not constitute a criminal act under the Criminal Code. To correct the disconnect between the way counterfeiting data are currently being collected and the objective of the UCR2 Survey, a change has been made to the way counterfeiting offences are counted.
The UCR Survey will now only count counterfeiting incidents submitted directly by police services and, more specifically, only those incidents where an accused person was identified. This would ensure that counterfeiting bills detected by financial institutions and those passed unwittingly by individuals would not be included in crime counts (Table 8).
This change, displayed in Chart 3.1 below, results in much lower counts of police-reported incidents of counterfeiting over the past 10 years. On average, the number of counterfeiting incidents dropped by about 97% in the years to which the adjustments were applied, from 1998 to 2007. The trend in the revised counterfeiting rate shows much greater stability over the past decade than did the previously published rate. The large increases originally published between 2001 and 2004 are no longer present in the revised counterfeiting rate.
Robbery
A change has also been made to the way in which robbery incidents are counted to bring this offence into line with the way in which other violent offences are counted. While all other violent crimes are counted according to the number of victims involved regardless of the number of distinct incidents that had occurred, robbery offences were counted according to the number of incidents that had taken place. For instance, if three people were assaulted at the same time and place by the same perpetrator, that would be counted as three assaults. However, if three people were robbed at the same time and place by the same perpetrator, that would be counted as only one robbery.
Until recently it was not possible to separate incidents where multiple victims were involved in a robbery from incidents where a number of people were simply present while a robbery occurred, such as in a retail store or a bank. Using the detailed UCR2 data now available, victims can be better identified. This means that robberies can be treated like all other violent crimes, with each victim counting as one robbery.
This change to robbery counts, illustrated in Chart 3.2, has resulted in an increase of about 12% annually in the number of reported robbery incidents for the period from 1998 to 2007. This has resulted in more robberies and a higher rate overall, but absolutely no change in the trend over the past 10 years.
Impact of these changes on the overall crime rate
To reflect the changes in offence counts, revisions have been made to the historical crime rate series dating back to 1998 1 and are illustrated in Chart 3.3. While some year-over-year impacts can be seen, such as in 2002 and 2004, the overall trend in the national crime rate between 1998 and 2007 is very similar after the revisions.
The revised rate shows a 7% drop in the police-reported crime rate over this period, while previously published data showed a 6% decline.
Reporting crime statistics
Now that virtually all police services in Canada are reporting to the UCR2 survey, more information than ever is available on crimes coming to the attention of police. This new information is being introduced into the standard data tables that Statistics Canada releases to the public each year.
These tables will feature more-detailed offence categories and improved crime categories. For example, the category of violent crime will now be broadened to include a number of distinct violent offences that previously had to be combined into an "other crimes" category, such as criminal harassment and uttering threats.
It is not uncommon for a number of criminal offences to occur at the same time and in the same place; in other words, as part of one incident. To avoid over-counting, crime rates are calculated based on the most serious offence in an incident. In the past, only the most serious offence in an incident was recorded by police.
Now, police services can send up to four different offences for each incident. This allows for a better understanding of the multiple types of offences that may occur in any one incident. As such, tables can be produced showing offence counts as only the most serious offence in an incident or as any offence in the incident.
The UCR2 survey also collects information on each victim in a violent incident. This allows for the counting of both the number of violent incidents reported by police and also the number of victims of violent incidents coming to the attention of police.
These improvements to reporting are further outlined in Appendix C of this report.
- Date modified: